huggingface/transformers

Pull Requests Commits

adding user defined tokens #30824

Ita Zaporozhets committed 2 years ago

83e3e1f5

Ita Zaporozhets committed 2 years ago

6ff31feb

added assertions

Ita Zaporozhets committed 2 years ago

3819ab54

moving split special token test to common tests

Ita Zaporozhets committed 2 years ago

3f3e7dc5

add new line at end of file

Ita Zaporozhets committed 2 years ago

7d4c047c

split special token

Ita Zaporozhets committed 2 years ago

6094454c

seems like `split_special_tokens` is used here

ArthurZucker committed 2 years ago

c578738a

FIX / TST: Fix expected results on Mistral slow test (A10) (#30909)

younesbelkada committed 2 years ago

Verified c876d121

[docs] Spanish translation of model_memory_anatomy.md (#30885)

aaronjimv committed 2 years ago

Verified 0df888ff

Add torch.compile for Mistral (#30642)

zhenglongjiepheonix committed 2 years ago

Verified 616bb11d

Introduce configured_state arg for accelerator_config (#29781)

muellerzr committed 2 years ago

Verified 92d1d97c

`tokenizer_class = "AutoTokenizer"` Llava Family (#30912)

ArthurZucker committed 2 years ago

Verified bb48e921

Fix a shape annotation and typos in `mamba` slow forward (#30691)

vasqu committed 2 years ago

Verified 76e05301

Add AutoFeatureExtractor support to Wav2Vec2ProcessorWithLM (#28706)

ylacombe committed 2 years ago

Verified e6708709

fix for custom pipeline configuration (#29004)

not-lain committed 2 years ago

Verified c11ac785

separate kwargs in processor (similar to #30193) (#30905)

Eric2i committed 2 years ago

Verified 7b4b4564

Fix num_hidden_layers in initialization of new model in Mamba (#30403)

SrGonao committed 2 years ago

Verified 18349164

add return_token_timestamps to WhisperProcessor (#30812)

kamilakesbi committed 2 years ago

Verified 1c2bb3ac

DeformableDETR two stage support bfloat16 (#30907)

DonggeunYu committed 2 years ago

Verified 66b0d9ee

LLaVa-Next: Update docs with batched inference (#30857)

zucchini-nlp committed 2 years ago

Verified 5d0bf59b

Add support for torch.compile dynamic shapes (#30560)

warner-benjamin committed 2 years ago

Verified cd6bd0af

FIX / Quantization: Fix Dockerfile build (#30890)

younesbelkada committed 2 years ago

Verified fce78fd0

Add TokenClassification for Mistral, Mixtral and Qwen2 (#29878)

josephenguehard committed 2 years ago

Verified 07bf2dff

Enable dynamic resolution input for Swin Transformer and variants (#30656)

the-neural-networker committed 2 years ago

Verified 481a9578

ArthurZucker committed 2 years ago

b6eb708b

Add fixed resize and pad strategy for object detection (#30742)

qubvel committed 2 years ago

Verified bf646fbf

update release script (#30880)

ArthurZucker committed 2 years ago

Verified e9a8041d

Support arbitrary processor (#30875)

ArthurZucker committed 2 years ago

Verified 0a9300f4

[whisper] fix multilingual fine-tuning (#30865)

sanchit-gandhi committed 2 years ago

Verified 57edd84b

Fix dependencies for image classification example (#30842)

jla524 committed 2 years ago

Verified 977ce58a

Older